How does the calculation work?
The calculator uses the standard formula for a proportion: n = z² × p × (1 − p) / e², where z is the confidence level (1.96 at 95%), p the expected proportion and e the margin of error. For a small population, such as your own customer base, the calculator applies a correction.
Common sample sizes
| Respondents | Margin of error (95%) | Suitable for |
|---|---|---|
| 300 | ±5.7% | A first exploration, one target group |
| 500 | ±4.4% | Most omnibus questions and concept tests |
| 1,000 | ±3.1% | National figures split by region |
| 2,000 | ±2.2% | Detailed segmentation, both language groups separately |
A large sample only helps if it is representative. Day One works with quotas on age, gender and region, and weights results where needed. Curious what your study costs? Try our cost calculator.
Frequently asked questions about sample sizes
Why use 50% as the expected proportion?
At 50% the required sample is largest. If you don't know how answers will split, it is always the safe choice.
Does a sample of 100 respondents make sense?
For a rough indication yes, but the margin of error is then about ±10%. For decisions we recommend at least 300 respondents.
What is the difference between margin of error and confidence level?
The margin of error says how far the result may be from reality. The confidence level says how sure you are that reality lies within that margin.